Import and process data

Learning

Model: Correct responses by continuous age, trial, block number, and block condition

	correct_response_made
Predictors	Odds Ratios	SE
age scaled	1.1540	0.0918
learning trial scaled	1.2940	0.0420
reward condition1	1.6592	0.1106
block number scaled	1.1736	0.0605
age scaled × learning trial scaled	1.0075	0.0325
age scaled × reward condition1	1.0787	0.0717
learning trial scaled × reward condition1	1.1362	0.0351
age scaled × block number scaled	1.0387	0.0534
learning trial scaled × block number scaled	1.0197	0.0229
reward condition1 × block number scaled	1.1224	0.0603
age scaled × learning trial scaled × reward condition1	0.9718	0.0298
(age scaled × learning trial scaled) × block number scaled	0.9782	0.0218
(age scaled × reward condition1) × block number scaled	1.0203	0.0547
(learning trial scaled × reward condition1) × block number scaled	1.0134	0.0254
(age scaled × learning trial scaled × reward condition1) × block number scaled	0.9767	0.0243
Random Effects
σ²	3.29
τ₀₀ _{subject_id}	0.20
τ₁₁ _{subject_id.re1.learning_trial_scaled}	0.02
τ₁₁ _{subject_id.re1.reward_condition1}	0.14
τ₁₁ _{subject_id.re1.block_number_scaled}	0.08
τ₁₁ _{subject_id.re1.learning_trial_scaled_by_reward_condition1}	0.02
τ₁₁ _{subject_id.re1.learning_trial_scaled_by_block_number_scaled}	0.01
τ₁₁ _{subject_id.re1.reward_condition1_by_block_number_scaled}	0.09
τ₁₁ _{subject_id.re1.learning_trial_scaled_by_reward_condition1_by_block_number_scaled}	0.01
ρ₀₁
ρ₀₁
ICC	0.06
N _{subject_id}	34
Observations	14688
Marginal R² / Conditional R²	0.105 / 0.157

Model: Correct response to first appearance of each stimulus

	correct_response_made
Predictors	Odds Ratios	SE
age scaled	1.0283	0.0597
category rep scaled	1.1410	0.0590
reward condition1	1.3145	0.0910
block number scaled	0.9974	0.0501
age scaled × category rep scaled	0.9612	0.0512
age scaled × reward condition1	1.0676	0.0740
category rep scaled × reward condition1	1.1810	0.0610
age scaled × block number scaled	1.0176	0.0514
category rep scaled × block number scaled	0.9513	0.0472
reward condition1 × block number scaled	1.0972	0.0551
age scaled × category rep scaled × reward condition1	1.0467	0.0557
(age scaled × category rep scaled) × block number scaled	1.1090	0.0583
(age scaled × reward condition1) × block number scaled	1.0949	0.0553
(category rep scaled × reward condition1) × block number scaled	1.0052	0.0499
(age scaled × category rep scaled × reward condition1) × block number scaled	0.9223	0.0484
Random Effects
σ²	3.29
τ₀₀ _{subject_id}	0.03
τ₁₁ _{subject_id.re1.reward_condition1}	0.08
ρ₀₁
ρ₀₁
ICC	0.01
N _{subject_id}	34
Observations	1836
Marginal R² / Conditional R²	0.047 / 0.055

Model: Category win-stay lose-shift

	WSLS
Predictors	Odds Ratios	SE
age scaled	1.0183	0.0373
learning trial scaled	1.0058	0.0181
reward condition1	1.3492	0.0469
block number scaled	1.0778	0.0194
age scaled × learning trial scaled	0.9382	0.0168
age scaled × reward condition1	1.0573	0.0366
learning trial scaled × reward condition1	1.1178	0.0202
age scaled × block number scaled	0.9941	0.0178
learning trial scaled × block number scaled	0.9921	0.0179
reward condition1 × block number scaled	1.0728	0.0193
age scaled × learning trial scaled × reward condition1	0.9967	0.0179
(age scaled × learning trial scaled) × block number scaled	0.9848	0.0177
(age scaled × reward condition1) × block number scaled	1.0471	0.0187
(learning trial scaled × reward condition1) × block number scaled	1.0160	0.0183
(age scaled × learning trial scaled × reward condition1) × block number scaled	0.9898	0.0178
Random Effects
σ²	3.29
τ₀₀ _{subject_id}	0.03
τ₁₁ _{subject_id.re1.reward_condition1}	0.03
ρ₀₁
ρ₀₁
ICC	0.01
N _{subject_id}	34
Observations	13814
Marginal R² / Conditional R²	0.036 / 0.046

Memory

Model: AUCs by age, reward condition, memory specificity

	AUC
Predictors	Estimates	SE
age scaled	0.0058	0.0158
reward condition1	-0.0185	0.0065
foil type1	0.0464	0.0065
age scaled × reward condition1	0.0063	0.0065
age scaled × foil type1	-0.0063	0.0065
reward condition1 × foil type1	0.0109	0.0065
age scaled × reward condition1 × foil type1	-0.0022	0.0065
Random Effects
σ²	0.01
τ₀₀ _{subject_id}	0.01
ICC	0.55
N _{subject_id}	34
Observations	136
Marginal R² / Conditional R²	0.178 / 0.631

RL modeling

Choice weights

Model: Choice weights by block condition and age

	est
Predictors	Estimates	SE
abstraction1	0.0563	0.0800
reward condition1	0.0574	0.0800
age scaled	0.3045	0.1415
abstraction1 × reward condition1	0.1009	0.0800
abstraction1 × age scaled	-0.0777	0.0803
reward condition1 × age scaled	0.0153	0.0803
(abstraction1 × reward condition1) × age scaled	-0.0266	0.0803
Random Effects
σ²	0.87
τ₀₀ _{subject_id}	0.46
ICC	0.34
N _{subject_id}	34
Observations	136
Marginal R² / Conditional R²	0.081 / 0.398

Model: Exemplar choice weights by condition

	est
Predictors	Estimates	SE
reward condition1	-0.0435	0.0814
Random Effects
σ²	0.45
τ₀₀ _{subject_id}	1.00
ICC	0.69
N _{subject_id}	34
Observations	68
Marginal R² / Conditional R²	0.001 / 0.689

Relations between choice weights and points earned

## Mixed Model Anova Table (Type 3 tests, S-method)
## 
## Model: total_points ~ age_scaled * beta_scaled * abstraction * reward_condition + 
## Model:     (1 | subject_id)
## Data: beta_ests_points
##                                                 Effect        df          F
## 1                                           age_scaled  1, 31.94       2.00
## 2                                          beta_scaled 1, 119.78    7.97 **
## 3                                          abstraction  1, 89.40       0.43
## 4                                     reward_condition  1, 88.01 165.86 ***
## 5                               age_scaled:beta_scaled 1, 120.00       0.06
## 6                               age_scaled:abstraction  1, 90.31       0.10
## 7                              beta_scaled:abstraction 1, 106.86       0.01
## 8                          age_scaled:reward_condition  1, 88.71       0.51
## 9                         beta_scaled:reward_condition  1, 94.50    9.92 **
## 10                        abstraction:reward_condition  1, 87.91       0.45
## 11                  age_scaled:beta_scaled:abstraction 1, 108.85       0.00
## 12             age_scaled:beta_scaled:reward_condition  1, 94.47       0.07
## 13             age_scaled:abstraction:reward_condition  1, 89.19       0.21
## 14            beta_scaled:abstraction:reward_condition  1, 94.05   10.91 **
## 15 age_scaled:beta_scaled:abstraction:reward_condition  1, 91.51       0.56
##    p.value
## 1     .167
## 2     .006
## 3     .513
## 4    <.001
## 5     .809
## 6     .757
## 7     .942
## 8     .476
## 9     .002
## 10    .502
## 11   >.999
## 12    .796
## 13    .649
## 14    .001
## 15    .458
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

Relations between learning and memory

Do points earned during learning relate to memory?

Model: AUC by points earned, reward condition, age

	AUC
Predictors	Estimates	SE
age scaled	0.0054	0.0177
points scaled	0.0356	0.0153
abstraction1	0.0404	0.0099
reward condition1	-0.0381	0.0121
age scaled × points scaled	-0.0123	0.0144
age scaled × abstraction1	-0.0117	0.0093
points scaled × abstraction1	-0.0059	0.0113
age scaled × reward condition1	0.0076	0.0112
points scaled × reward condition1	0.0052	0.0145
abstraction1 × reward condition1	0.0141	0.0099
age scaled × points scaled × abstraction1	-0.0103	0.0103
age scaled × points scaled × reward condition1	-0.0142	0.0128
age scaled × abstraction1 × reward condition1	0.0029	0.0093
points scaled × abstraction1 × reward condition1	0.0111	0.0113
age scaled × points scaled × abstraction1 × reward condition1	0.0079	0.0103
Random Effects
σ²	0.01
τ₀₀ _{subject_id}	0.01
ICC	0.55
N _{subject_id}	34
Observations	136
Marginal R² / Conditional R²	0.235 / 0.659

Do choice weights relate to memory?

Model: AUC by age, exemplar choice weights, specificity, block condition

	AUC
Predictors	Estimates	CI
age scaled	-0.0019	-0.0343 – 0.0306
beta scaled	0.0171	-0.0075 – 0.0417
abstraction1	0.0509	0.0375 – 0.0642
reward condition1	-0.0163	-0.0297 – -0.0029
age scaled × beta scaled	-0.0103	-0.0334 – 0.0128
age scaled × abstraction1	-0.0108	-0.0246 – 0.0030
beta scaled × abstraction1	0.0057	-0.0081 – 0.0194
age scaled × reward condition1	0.0022	-0.0123 – 0.0168
beta scaled × reward condition1	0.0037	-0.0113 – 0.0187
abstraction1 × reward condition1	0.0129	-0.0004 – 0.0262
age scaled × beta scaled × abstraction1	-0.0131	-0.0248 – -0.0013
age scaled × beta scaled × reward condition1	-0.0038	-0.0161 – 0.0086
age scaled × abstraction1 × reward condition1	-0.0067	-0.0205 – 0.0072
beta scaled × abstraction1 × reward condition1	0.0044	-0.0094 – 0.0181
age scaled × beta scaled × abstraction1 × reward condition1	-0.0042	-0.0160 – 0.0075
Random Effects
σ²	0.01
τ₀₀ _{subject_id}	0.01
ICC	0.55
N _{subject_id}	34
Observations	136
Marginal R² / Conditional R²	0.214 / 0.649

AUC by exemplar choice weights: model effects

Model: AUC by age, category choice weights, specificity, block condition

	AUC
Predictors	Estimates	SE
age scaled	0.0034	0.0157
beta scaled	0.0161	0.0106
abstraction1	0.0464	0.0064
reward condition1	-0.0185	0.0066
age scaled × beta scaled	-0.0133	0.0105
age scaled × abstraction1	-0.0071	0.0065
beta scaled × abstraction1	0.0111	0.0065
age scaled × reward condition1	0.0048	0.0066
beta scaled × reward condition1	0.0136	0.0073
abstraction1 × reward condition1	0.0115	0.0064
age scaled × beta scaled × abstraction1	-0.0008	0.0061
age scaled × beta scaled × reward condition1	-0.0117	0.0068
age scaled × abstraction1 × reward condition1	-0.0028	0.0065
beta scaled × abstraction1 × reward condition1	0.0003	0.0065
age scaled × beta scaled × abstraction1 × reward condition1	-0.0111	0.0061
Random Effects
σ²	0.01
τ₀₀ _{subject_id}	0.01
ICC	0.56
N _{subject_id}	34
Observations	136
Marginal R² / Conditional R²	0.237 / 0.664

Memory Specificity Analyses: Experiment 2 - Children Only, Continuous Age Analyses

Kate Nussenbaum

6/13/24

Import and process data

Learning

Model: Correct responses by continuous age, trial, block number, and block condition

Model: Correct response to first appearance of each stimulus

Model: Category win-stay lose-shift

Memory

Model: AUCs by age, reward condition, memory specificity

RL modeling

Choice weights

Model: Choice weights by block condition and age

Model: Exemplar choice weights by condition

Relations between choice weights and points earned

Relations between learning and memory

Do points earned during learning relate to memory?

Model: AUC by points earned, reward condition, age

Do choice weights relate to memory?

Model: AUC by age, exemplar choice weights, specificity, block condition

AUC by exemplar choice weights: model effects

Model: AUC by age, category choice weights, specificity, block condition

Figure 5F: AUC by category choice weights: model effects